Project 1: Classification and Definition of Learning Disabilities Based on Response to Instruction Models. The objective of Project 1 (Classification) is to systematically evaluate a set of classification and measurement issues that affect the identification of students as learning disabled (LD) in response to instruction (RTI) models. We propose three specific aims organized around the classification, measurement, and identification of LD. Specific Aim 1 (Assessment and Identification) addresses the development of an assessment battery that permits the identification of poor readers in the older grades. To implement RTI models, it must be possible to screen a school population, identify students at-risk for disability status, and monitor their progress. In Year 1, we propose a study of 800 students followed from the end of Grade 5 - beginning of Grade 6 in an effort to (a) identify the incidence and types of reading problems involving accuracy, fluency, and comprehension in a sample initially selected on the basis of poor performance on a comprehension test;(b) evaluate the stability of these types;and (c) determine whether abbreviated assessments at ah individual or group level can reliably identify poor readers in middle school. Specific Aim 2 (Reliability of RTI Models) addresses the nature and reliability of different criteria for identifying inadequate responders in different assessment domains (accuracy, fluency) and criteria (intercept, slope, absolute criteria). By utilizing the assessment data used to evaluate outcomes in Projects 2 (Early Intervention) and 3 (Remediation) at the beginning and end of intervention, and more frequent progress monitoring assessments from these Projects, we will define the number of students that meet different RTI criteria, examine overlap, and manipulate the effect of different cut points. Specific Aim 3 (External Validity of RTI Models) addresses the validity of classifications that emerge under Aim 2 using cognitive assessments obtained before and after Tier III interventions in Project 2 (Early Intervention) and Project 3 (Remediation). In addition, we will use the functional neuroimaging data from Project 4 (MSI) to further evaluate the external validity of RTI classifications based on variables not used to define the groups. RTI criteria will be compared to criteria generated by other proposed classification, including IQ-discrepancy, low achievement, and individual difference models. Altogether, we propose a thorough investigation of classification and definition issues involving LD and RTI that reflects our 25 year history of research on these issues and is integrated with the other projects and cores in the Texas Center for Learning Disabilities.